Application Research Of Fault Diagnosis For Shearer Based On Fuzzy Decision Tree

Coal mine safety production is the basis and guarantee of sustainable development in aspect of coal economy. Because the coal mine accidents happen frequently, how to improve the safety and reliability of coal mine machinery equipment becomes one universal attention question. Shearer is a key equipment in coal production, whose fault diagnosis has been gradually paid attention to. How to timely make an accurate fault diagnosis of shearer to improve the use rate of it is a common research topic.On the basis of in-depth study of the shearer fault diagnosis methods at home and abroad, comparing and analyzing the shortcomings of the methods that are fault diagnosis methods of shearer, such as neural network, support vector machine, discussing the merits of fuzzy decision tree, this paper applies fuzzy decision tree in the fault monitoring and diagnosis. According to the characteristics of fuzzy decision tree structure related to heuristic strategies, the dependency degree which rough set has is used as a heuristic method to generate fuzzy decision tree, thus an improved algorithm for fault diagnosis is proposed. The algorithm can quickly and accurately diagnose the fault of the shearer.Aiming at the problem that is the low efficiency of fuzzy c-means clustering algorithm obtaining the fuzzed sample data, an improved fuzzy c-means clustering algorithm is proposed. The algorithm introduces the idea of gravity, regards data objects as the physical particles, and considers the macro forces'impact on clustering results between a number of data objects. It can solve the problems that are high time complexity and slow convergence of the distance-based fuzzy c-means clustering algorithm. In this paper, after the improved fuzzy c-means clustering algorithm is used to preprocessing the shearer failure data, improved fuzzy decision tree algorithm is used to classify fuzzed data to obtain fuzzy classification rules. These rules provide auxiliary decision support for shearer fault diagnosis decision-makers. Experiments show that the method improves the efficiency and accuracy of shearer fault diagnosis, which provides a reliable basis for the shearer fault diagnosis quickly and efficiently.